8/2/2019 Hum. Reprod.-1996-Wyss-1992-7

1/6

Human Reproduct ion vol.11 no.9 pp 1992-1997, 1996

Regeneration processes in rabbit endometrium: aphotodynamic therapy model

P.Wyss1*2, R.Steiner1 '2, L.H.Liaw1, M.T.Wyss1-2,A.Ghazarians1 , M.W.Berns1, B J .Tromberg 1 an dY.Tadir1'3'Beckman Laser Institute and Medical Clinic, Departments ofObstetrics and Gynecology, University of California, Irvine, CA,USA and 2Umversity of Zurich, Switzerland^ o whom correspondence should be addressedThe origin and process of regeneration in rabbit endo-metrium was evaluated following photodynamic epithelialdestruction using topically applied aminolevulinic acid(AL A). Selective destruction of endometrial epithelium wasperformed using photodynamic therapy (PDT). ALA wasdiluted to 200 m g/ml dextran 70 shortly prior to administra-tion. A volume of 1.2 ml was injected into the left uterus.Intrauterine illumination (wavelength 630 run, light dose40-80 J /cm 2) was p erformed 3 h after drug administration.Tissue morphology was evaluated by light and scanningelectron microscopy 1, 3, 7 and 28 days post-treatment(three animals at each time-point). Regeneration of theendometrium following epithelial ablation by PDT wasfully activated after 24 h and was completed after 72 h.Endometrial surface generation occurred by proliferation,originating primarily in deeper regions of the glands.Findings from our morphological follow-up study supportthe origin of endometrial regeneration being mainlyfrom undifferentiated stem cells and residual glandularepithelium.Key words: endometrium/photodynamic therapy/regeneration

IntroductionThe regenerative capacity of the endometrium is regarded asunique and as one of the most dynamic phenomena in humans.It is characterized by cyclic proliferation, differentiation andcell death every menstrual cycle. The duration of menstrualsurface re-epithelialization is ~48 h and begins on cycle days2 or 3 (Ferenezy, 1977). Tissue regeneration at comparablerates is known only in the haematopoietic (Lajtha, 1973),intestinal (Hagem ann and Lesher, 1973) and epidermal systems(Lavker and Sun, 1983). Several morphological studies onendometrial regeneration in humans (McLennan and Rydell,1965; Baggish et al, 1967; Ferenezy, 1976a,b) and rabbits(Schenker et al, 1971; David et al, 1973; Beier and Mootz,1978) have already demonstrated its regenerative potential.

To understand the biological principles on which quicklyrenewing tissues are based, several hypotheses have beenoffered: (i) a small pool of multipotent 'undifferentiated stem1992

cells' located near or within the endometrialmyometrialjunction give rise to epithelial, stromal or vascular cells(Cairnie etal, 1976; Prianishnikov, 1978); (ii) the regen eratingsurface epithelium originates from the residual epithelium ofgland stumps or undamaged bordering epithelium (Bartelmez,1933; McLennan and Rydell, 1965); (iii) stromal cells trans-form to endometrial epithelium, thereby relining the surface(Papanicolaou, 1933; Craig, 1963); and (iv) rupture ofcapillaries gives rise to migration and transformation of theendothelial cells.

Photodynamic therapy (PDT) offers a new approach to thestudy of endometrial epithelial regeneration because it offersthe selective targeting and destruction of cells within theendometrium. The technique typically involves the i.v. ortopical administration of a photosensitizing drug. When lightof sufficient energy and appropriate wavelength interactswith the sensitizer, highly reactive oxygen intermediates aregenerated (Kimel et al, 1989). These intermediates, primarilysinglet molecular oxygen, irreversibly oxidize essential cellularcomponents. The resulting photodestruction of crucial cellorganelles and vasculature ultimately causes cell necrosis. Inthis study, aminolevulinic acid (ALA) was used to target theendometrium. ALA is a precursor of protoporphyrin EX (PpIX)in the biosynthetic pathway of haemoglobin. Haemoglobinbiosynthesis is essential to life and occurs in all aerobic cells.

The slowest step in this process is the conversion of PpIXto haemoglobin. Therefore, the administration of exogenousALA induces the accumulation of PpIX, a strong photosensit-izer. Because only certain types of cell have a large capacityto localize PpEX, the use of ALA for PDT provides potentialselectivity. After the intrauterine application of ALA, PpDChas been found to be metabolized more readily in endometrialglands than in the surrounding stroma (Wyss et al, 1994b).Therefore, selective targeting and damage of the endometrialepithelium is possible, especially by using light doses underthe threshold of complete and permanent endometrial destruc-tion. The aim of this study was to test several hypotheses forthe origin of endometrial regeneration by using PDT to observethe regenerative process of rabbit endometrium followingepithelial ablation. If successful, diis concept might offer anew approach to the study of endometrial physiology.

Materials and methodsAnimalsA total of 15 mature female New Zealand White rabbits (n = 15)weighing 3600-4300 g were placed in a controlled environment withfree access to food and water. Three rabbits were killed to evaluatebistological changes at 1, 3, 7 and 28 days following PDT after

European Society for Human Reproduction and Embryology

8/2/2019 Hum. Reprod.-1996-Wyss-1992-7

2/6

Pbotodynamic therapy and the study of endometrial regeneration

Fig ure 1. Scanning electron micrographs (original magnification X25 2) (A) An untreated endom etnal surface with muco sal folds (mf) andgland openings (go) at oestrous stage. (B) The same aspect 24 h after photodynamic therapy The columnar surface epithehum is missing,while the gland openings are still visible.

intrauterine ALA instillation and 28 days after uitrauterine benzopor-phynn derivative administration. Rabbits are induced ovulators,therefore no oestrous monitoring was required.PhotosensitizerCrystallized 5-ALA (DUSA USA Inc., Parsippany, NJ, USA) andvials of hposomally formulated benzopo rphyrin denva uve MA (BPD)(Quadra Logic Technologies Inc., Vancouver, British Columbia,Canada) were stored in the dark at a temperature of 4C. ALA wasdiluted to 200 mg/ml and BPD to 2 mg/ml in Hyskon (dextran 70;Kabi Pharmaceuticals Inc, Clayton, NJ, USA) shortly prior toadministration. To minimize acidity, ALA/Hyskon solutions weretitrated with 10 and 1 N NaO H to pH 5.5 . The injected volume w as1 2 ml in rabbit uteri (left one only)ProceduresAnimals were anaesthetized with 0 75 ml/kg ketamine/xylazine (2 1)l.m., and isofluorane was added during the surgical intervention.Intrautenne drug application was performed through a midlineincision. The photosensitizer was injected with a 20 gauge needleinto the uterus 3-5 mm distal to the uterine bifurcation. Abdominalwalls were closed in three layers [Dexon 4-0 (Davis & Geek, Wayne,NJ, USA) and staples].

At 3 h after ALA application and 1 5 h following BPD administra-Uon, re-laparotomy was performed. Light from an argon-pumped dyelaser operating at 630 nm for ALA and 690 nm for BPD (SpectraPhysics, Mountain View, CA, USA) was delivered into the uterinecavity via a 400 mm diameter quartz optical fibre terminated witha 3.0 cm long cylindrical diffusing tip (model 4420-A02, PDTSystems, Buellton, CA, USA). A clinical Hartridge reversionspectroscope (Ealing Electro-Optics, South Natick, MA, USA) wasused to verify the wavelength. Because the length of the rabbit uteruswas 12-15 cm, multiple (four or five) incremental irradiations wererequired. A total of 185 mW was launched into the fibre (65 raW/cm fibre tip) during 800 s, resulting in a variable ussue dose which,depending on geometry, ranged from 40 to 80 J/cm 2.Specimen retrievalThe rabbits were first anaesthetized with isofluorane and then under-went euthanasia by intracardiac injection of 1 5 ml Euth-6, a barbituricacid derivative/central nervous system depressant used only foreuthanasia of animals (Western Medical Supply, Arcadia, CA, USA)

Uten were retrieved immediately following euthanasia. Specimenswere sectioned in four blocks of 34 mm each and fixed in 10%formaldehyde.Plastic-embe dded section for light microscopySamples were fixed in Kamovsky's fixative (2% paraformaldehyde,3% glutarald ehyd e in 0 1 M cacodylated buffer) for 2 4 h at roomtemperature and then nnsed in a 0 1 M cacod ylated buffer. Thesamples were then post-fixed in 1% asonium tetroxide in 0.1 Mcaco dylate buffer for 1 h, nns ed with doub le-distilled water andstained en bloc in Kellenbeger's uranyl acetate for 2 h. Dehydrationwas performed with progressive concentrauons of ethanol-water in10 mm steps (30, 50, 70, 80 and 100%) and also in progressiveetha nol-p ropy leno xide in 10 min steps. Infiltration was performedwith progressive propylenoxide (Poly/Bed 872; Polysciences,Warnn gton, PA, USA) in 30 min steps Each sample was embeddedcarefully in flat mold. Sections, 500 nm thick, were cut using ahistodiamond knife (Diatome US, Fort Washington, PA, USA) on anultramicrotome and stained with Richardson's stainScanning electron microscopy (SEM )Samples were fixed in 10% formalin in phosphate buffer at roomtemperature for 24 h, post-fixed in 10% osmium tetroxide, dehydratedm graded acetone cnucal point-dned (Ladd critical point dryer, LaddResearch Industries Inc., Burlington, VT, USA) and sputter coatedwith gold (Pelco PAC-1 evaporating system; Ted Pello Inc , Redding,CA, USA ) Micrographs were taken on a SEM (SEM 515, PhilipsElectronic Instrument Company, Mahwake, NJ, USA)

ResultsThe morphological changes following selective epithelialdamage (glands and luminal epithelium) in rabbit en dom etnumby photochemical effects are demonstrated in a sequence ofimages. The luminal surface of an untreated uterus showedmucosal folds covered by polygonal bordered (columnar)epithelial cells and several gland openings (Figure 1A). Thelight microscopy image of a transverse aspect of the endo-metrial structures (Figure 2A) exhibited a columnar epitheliumsurfacing the lumen and a gland. The underlying stromaconsisted of loosely structured fibroblasts with extracellular

1993

http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/

8/2/2019 Hum. Reprod.-1996-Wyss-1992-7

3/6

Figure 2. Light microsco pic im ages (A and B, original magnification X790 ); scanning electron micrograph (C , original m agnificationX2020). (A) Transverse aspect of an untreated uterus (endometrium) exhibiung luminal (le) and glandular columnar epithelium (ge), stromawith fibroblasts (fb) and ex tracellular matrix (em). (B) Same endom etrial aspect 24 h after photodynam ic therapy (PD T). Columnar surfaceepithelium is absent. Round, flattened cells [see arrows, are protruding out of the gland openings (go)]. Cellular compactness in the stromais decreased, whereas extracellular matrix is more abundant. (C) Chain of recovering flattened cells (ch) at a gland opening (go) (24 h afterPD T)

matrix (collagen fibres, proteoglycan-glycoprotein ground sub-stances) and capillaries. Stroma] cellular density was highernear the uterine lumen than in the deeper endometrial regions.Myometrium (not shown) consisted of an internal circularand an external longitudinal muscle layer, and displayed nomorphological change after PDT.Endometrial morphology changed completely 24 h afterPDT. In these specimens, mucosal folds were flattened, theepithelial layer was absent but the gland openings were still

visible (Figure IB ). A few spherical cells were located primarilyin and around the gland openings. The transsection of a gland(Figure 2B) 24 h after treatment showed flattened 'spindle'cells covering the luminal surface. Columnar epithelium wasabsent in these cells and their contour was more rounded atthe gland opening. Cellular density in the stroma was reduced,while extracellular matrix was more abundant. Fibroblastswere directed towards the luminal surface. The superficialaspect of a similar gland opening (Figure 2C) displayed achain of these flattened epithelial cells protruding out of thegland. These cells appeared to be more spherical when distant1994

from the gland opening. The subjacent ground may representthe basal membrane. Figure 3A and B demonstrates themorphological difference between untreated luminal epitheliumand the migrating regenerative epithelial cells in high magni-fication (SEM m agnification X500 0). Normal oestrous endo-metrial epithelium consisted of ciliated cells surrounded bynon-ciliated microvillous (secretory) cells. Regenerating epi-thelial cells exhibited a smooth surface with scant cilia ormicrovilli and appeared to lack functional specialization.In a few of the images taken at 24 h after PDT, slightlypale cells containing large pale nuclei with irregular chromatincould be observed at the luminal surface (Figure 4). In thesecells the nucleoli were not evident and the cell borders werepoorly defined. The morphology of these cells was moresimilar to fibroblasts than to epithelial cells. Underlying stromahad abundant extracellular matrix, and stromal cells weredirected towards the fibroblast-like cells which were likely tobe participating in the surface regeneration process.Capillaries were sometimes very close to the epithelialstructures of the endometrium. In a small number of images

http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/

8/2/2019 Hum. Reprod.-1996-Wyss-1992-7

4/6

Photodynamic therapy and the study of endometrial regenerat

Figure 3. Scanning electron micrographs (original magnification X5000). (A) Untreated endometrial surface (oestrous stage) with a ciliatecell (cc) surrounded by microvillous (secretory) cells (me). (B) Flattened re-epithelializing cells (re) containing thread-like structures (ths)communicating with each other and with the base (24 h after photodynamic therapy).

Figure 4. Light microscopic image (original magnification X790).Slightly pale fibroblast-like cells (see arrows) with large pale nucleiwere presumably participating in regeneration of the endometrialsurface (es) (24 h after photodynamic therapy).

taken 24 h post-treatment, endothelial cells were separated byonly a thin layer of extracellular matrix and the basal m embranefrom the de-epithelialized endometrial surface (Figure 5). Adirect participation of endothelial cells in the re-epithelializingprocess could not be detected microscopically.

By 3 days after photodynamic damage of the epithelialstructures using relatively low light levels, complete recoveryof the endometrial surface was observed (data not shown).Ciliated cells were also distinguishable from microvilloussecretory cells, but there were less mucosal folds visible. At1 and 4 weeks after PDT, endometrial folds were reconstructedand the regenerated endometrium was identical to theuntreated one.In contrast, complete loss of the epithelial structures was

observed in uteri 4 weeks following PDT using BPD as aphotosensitizer. The surface was replaced by a collagen n etworkresembling scar tissue (Figure 6).

Figure 5. Light microscopic image (original magnification X790)Some images exhibited endothelial cells of capillaries (see arrowsseparated only by a thin layer of extracellular matrix from the basmembrane (bs) and the endometrial surface (es) (24 h afterphotodynamic therapy). Their morphological participation in theregenerative process could not be documented microscopically.DiscussionThe endometrium of humans and mammals displays astonishing regenerative capacity (Hartman, 1944; McLennand Rydell, 1965; Schenker et al, 1971; Ferenezy, 197Prianishnikov, 1978; Kaiserman-Abramof and Padykula, 198Ludwig et al, 1990). This capacity is based on essentintrinsic endometrial tissue mechanisms that are independen tthe hormonal influences of the reproductive system (Ferenez1976a,b). Several hypotheses are proposed to explain tcellular origin of endometrial regeneration. This study wdesigned to use the principles of PDT to elaborate on tregenerative mechanism of the endometrium with suboptimlight doses for reversible selective destruction of the epithelstructures.At 24 h after PDT, our microscopic examination demostrated significant damage with the initial phase of regenerati

http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/

8/2/2019 Hum. Reprod.-1996-Wyss-1992-7

5/6

P.W y s s et al

10m20.0kU 568E3 2481/00 BLI SEM

Figure 6. Scannuig electron micrograph (original magnificationX 5000) Complete loss of luminal columnar epithelium andglandular openings 4 weeks following photodynamic therapy usingbenzoporphyrin derivative and sufficient light dose. The surfaceappeared to be replaced by a collagen network resembling scartissue or perhaps basal mem brane [Reprinted with permission(Wyss et al, 1994a).](Figure 2B and C). At this stage, most of the recovering cellswere located at the gland opening. These observations indicatethat endometnal surface generation after epithelial destructionoccurs by proliferation originating primarily in the glands.Several studies of endometrial regeneration following menstru-ation, abortion, and mechanical and pharmacological destruc-tion have suggested that the regenerating surface epitheliumoriginates from the residual glands and epithelial structures(Bartelm ez, 1933; Mc Lenn an and Rydell, 1965; Ferenezy,1977; Beier and Mootz, 1978). However, endometrial stemcells for epithelial, stromal and vascular components aresupposed to be located near to or within the endometrial-myometrial junction where the basal regions of the endometrialglands interdigitate with the myometrium (Padykula, 1989). Asmall pool of these undifferentiated multjpotenUal cells isassumed to be the main source of the endometrial regenerativepotency (Padykula, 1989).Interestingly, the post-ovulatory epithelial mitotic activityin the deeper endometrial regions (called basalis IV) escapesinhibition by progesterone and may already be prepared forregenerauve activity during menstruation (Kaiserman-Abramofand Padykula, 1989). Studies of the gastrointestinal mucosa(Hagemann and Lesher, 1973) and the epidermis (Lavker andSun, 1983) have suggested that stem cells are more likely tobe localized in the crypts. The location of these tissues inregions deeper than the lumen provides some protection andmay also maintain die ability to regenerate after injury. Theidentification of stem cells in the endometrium is difficultbecause they have not been characterized clearly.Other investigators have advocated a stromal origin ofend om etna l re-epithelialization (Papanicolaou, 1933; Baggishet al, 1967). The main argument supporting the hypothesis oftransformation of the stromal cells to epithelial cells is thecommon embryological origin of both cell types. Both celltypes are presumed to arise from the intra-embryonic mesodenn1996

that is built up by mesenchymal cells (Hinrichsen, 1990). TheMullerian (paramesonephric) system is developed witiin andby the mesodenn. This process includes the appearance of theintra-embryonic coelom, a cavity m the mesoderm covered bythe mesothelial cells, and the invagination of coelomic surfacebuilding up the Miillerian duct. The duct consists of coelomicmesothelium (epithelial) surrounded by mesodermal mesen-chyme. The mesothelium generates the endometrial epitheliumand the mesenchyme generates the endometnal stroma andmyometrium. Interestingly, the differentiation into endometnalepithelium is specified by mesenchyme (Cunha and Fujii,1981). Because stromal fibroblasts are able to degrade collagenand elastic fibres, cellular migration through extracellularmatrix and basal membranes should be possible. Our lightmicroscopy images show cells relining the endometrial surface,which correspond morphologically to fibroblasts (Figure 4).Re-epithelializing cells are often described as 'fibroblastoid'or 'spindle' cells (Baggish et al, 1967). However, electronmicroscopic studies could not support the hypothesis thatsurface epithelial cells may be denved from stromal fibroblasts(Ferenezy, 1976a,b). In addition, DNA synthesis in the stromalfibroblasts during the period of repair did not deviate from thenormal values (Ferenezy, 1977), except by a short increaseduring maximum epithelial regenerative activity 48 h afterinjury. This suggests that the transformation of fibroblasts forepitnelial regeneration is unlikely.The first embryonic blood vessels arise in die extra-embryonic mesenchyme (Hamilton and Mossman, 1972)Intra-embryonic mesenchyme (mesodermal cells) is supposedto generate embryonic vessels as well (Reagan, 1917). Mesen-chymal cells form angioblasts, which are generating endotnelialcells. Therefore, the endothelium, the stroma and the epitlieliumof the endometrium may have the same mesodermal origin.Endometrial destruction with vascular damage (haemorrhagic)activates endotnelial regeneration. Based on embryo develop-ment, precursors of endothehal cells may be transformed intostromal cells and be involved in the relining process of theendometnal epithelium. Capillaries were located just below

the basal memb rane (Figure 5) A direct participation ofendothelial cells in epithelial regeneration could not beobserved in our microscopic study.Regeneration of the endometrium following selectiveepitnelial destruction by PDT was fully activated at 24 h andcompleted after 72 h, resembling that of the non-treatedcontrol. In anodier study, following curettage of the rabbitendometnum, regeneration of the luminal epithelium wasrapidly initiated at 3 h and completed at 72 h (Schenker et al,1971). Regeneration of human endom etrium starts imm ediatelyafter the onset of menstrual bleeding and a continuous layerof fusiform cuboidal epimelial cells is produced up to day 6(Ludwig and Metzger, 1976).In conclusion, the regenerative capacity of endometrialepithelium may originate in different cell types. Multipotentundifferentiated stem cells, residual endometnal epithelium,stromal fibroblasts and endothehal cells of ruptured capillariesare ah" possib le participants in tins regeneration.Interestingly, regeneration of the epithelial structures didnot occur using the BPD as a photosensitizer (Wyss et al..

http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/

8/2/2019 Hum. Reprod.-1996-Wyss-1992-7

6/6

1994a) at the same light dose as for ALA. The absorptionpeak of BPD at 690 nm may offer a longer wavelength, withthe deeper penetration depth of light in human endometriumincreasing the light dose at the endometrial-myometria] con-junction where stem cells are supposed to be located. Anaugmented phototoxicity of BPD to regenerating cells mayoptimize photodynamic efficacy as well.The genuine mechanism of endometnal regeneration isdifficult to determine by microscopic investigations of cellmorphology. Our investigation was limited to the histologicallocalization of the definitive origin of regeneration in thedeepest portions of the endometrium. Even functional studiesmay not define the final mechanism (Ferenezy, 1977; Padykula,1989). Based on the embryological background, it is morelikely that several processes may participate in regenerationof the endometrium, and none of the aforementioned hypo-theses may be definitely excluded. Our morphological observa-tions support the assertion that endometnal regenerationoriginates in undifferentiated stem cells and residual g landularepithelium.Recent studies in our laboratories have demonstrated thatdifferent photosensitizers can target specific areas such as theendometrial glands or the stroma (Wyss et al., 1994a,b).Various combinations of drugs and light may offer a newapproach to the study of endometrial regeneration. Similarconcepts may be applicable for studying mechanisms of embryoimplantation, as demonstrated in our previous study in whichno visible histological damage at implantation failure followingPDT was demonstrated in rats (Sterner et al, 1995).

AcknowledgementsThis work was supported by grants from the National Institutes ofHealth (nos. 2RO1 CA32248 and 5P41 RR01192), the Departmentof Energy (no. DE-FG03-91ER61227) and the Office of NavalResearch (no. N00014-91-C-0134), a Memorial Health ServicesGrant, Krebshga of Switzerland and Academic Nachwuchsfoerderung,University of Zurich, Switzerland

ReferencesBaggish, M.S., Pauerstein, C J and Woodruff, J.D (1967) Role of stroma inregeneration of endometnal epithelium Am. J O bstet GynecoL, 99,459-465Bartelmez, WG (1933) Histological studies on the menstruating mucousmembrane of the human uterus Conxnb Embryol, 24, 141-148Beier, HJvl. and Mootz, U. (1978) Significance of maternal utenne proteinsin the establishment of pregnancy Ciba Found. Symp., 64, 111-140Cairnie, AJ3., Lala, P.K. and Osmond, D G (1976) Stem cells of renewingcell populations Academ ic Press, New York, NY, USA.Craig, J M (1963) Reoculum and collagen in the human endo metrium . Am .J. Obstet GynecoL, 86, 421-427Cunha, G.R and Fujii, H (1981) Stromal-parenchvm al interactions in normaland abnormal development of the genital tract In Herbst, A.L. and Bern,H.A (eds), Development Effects of Diethylsnlbestrol (DES) in PregnancyThieme-Stratton, New York, NY, USA, pp. 179-193David, A., Kaplun, D., SCIT, D M and Czemobilsky, B (1973) Effect ofintrautenne contraceptive device on the regeneration of rabbit endometnum.Am. J Obstet. GynecoL, 117, 473-477Ferenezy, A (19 76a) Studies on the cytodynam ics of human e ndotne tnalregeneration. I Scanning electron microscopy Am. J Obstet, GynecoL,124, 64-74Ferenezy, A. (1976b) Studies on the cytodynamics of human endometnalregeneration II Transmission electron microscopy and histochemistry Am .J Obstet. GynecoL, 124, 582-595

Photodynamic therapy and the s tudy of endometr ial regenerat ionFerenezy, A. (1977) Studies on the cytodynamics of expenme naal endometnalregeneration in the rabbit. I Historadioautography and ultrastructure Am .J Obstet GynecoL, 128, 536-546Hagem ann, R.F and Lesher, S. (1973) Intestinal cytodynamics addictionsfrom drug radiation studies In Zimme rman, A M and Padilla, B M (eds).Drugs and the Cell Cycle. Academic Press, New York, NY, USAHamilton, WJ. and Mossman, HW. (1972) Human Embryology Williamsand Wilkins, Baltimore, MD, USA.Hartman, C.G. (1944) Regeneration of the monkey uterus after surgicalremoval of the endometnum and accidental endometnosis West J Surg

Obstet GynecoL, 52, 87-102Hmn chsen, K.V (1990) Human -Embryolog ie EntwickJung der innerenGenitalorgane. Springer, Berlin, GermanyKaiserman-Abramof, I.R. and Padykula, H A (1989) Ultrastructure zonationof the pnmate endometnum (rhesus monkey) Am. J Anat, 184, 1330Kimel, S., Tromberg, BJ., Roberts, W G. and Bems, M W (1989) Singletoxygen generation of porphyrins, chlonns and phthalocyanines Photochem.PhotobioL, 50, 175-183Lajtha, LJ (1973) Review of leukocytes NatL Cancer Inst. Monogr , 38,111-123.Lavker, R.M and Sun, T (1983) Epidermal stem cells J Invest D ermawL,81 , 121-127Ludwig, H. and Metzger, H (1976) The re-epithehahzalion of end ometnu mafter menstrual desquamation. Arch. GynakxoL, 221, 51-60 .Ludwig, H , Metzger, H. and Frauli, M (1990) Endom etnum . tissue remodelingand regeneration. In D'Ac angu es, C , Fraser, I. S , Newton, J.R. andOdhnd, V (eds), Contraception and Mechanisms of Endometnal BleedingCambndge University Press, Cambridge, MA, USAMcLennan, C and Rydell, A (1965) Extent of endometnal shedding dunngnormal menstruation. Obstet GynecoL, 26, 605-621.Padykula, H A (1989) Regeneration in the pnm ate uterus: the role of stemcells In Wynn, R.M and Jolhe, WP . (eds), Biology of the Uterus PlenumMedical Book Co., New York, NY, USA, p. 279Papanicolaou, G (1933) Epithelial regeneration in the utenne glands and onthe surface of the uterus Am. J Obstet G ynecoL, 25, 30-37Pnanishrukov, VA. (1978) On the concept of the stem cell and a model offunctional morphological structure of the endometnum Contraception, 18 ,213-223Reagan, F.P (1917) Expenmental studies on the ongin of vascular endotheliumand of erythrocytes. Am. J. AnaL, 3998Schenker, J.G , Sacks, M L and Polishuk, W 2 (1971) Regeneration of rabbitendometnum following curettage. Am. J. Obstet GynecoL, 111, 970978Sterner, R , Tromb erg, B , Wyss, P et aL (1995) Rat reproductive performancefollowing photodynamic therapy with topically administered photofnnHum. Reprod., 10, 227-233.Wyss, P, Tadir, Y, Tromberg, BJ et al (1994a) Benzoporphynn denvative(BPD)' a potent photosensitizeT for photodynamic destruction of the rabbitendometrium. Obstet GynecoL, 84, 409-414Wyss, P., Tromberg, BJ., Wyss, M T et aL (1994b) Photodynamic destructionof endometrial tissue using topical 5-aminolevuhnic acid (5-ALA) in ratsand rabbits Am J Obstet Gynecol, 171, 1176-1183

Received on January 16, 1996, accepted on June 19, 1996

1997

http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/http://humrep.oxfordjournals.org/